Rebellion in the

Internet Age

Figure 1: A comic highlighting the powerful role of social media.1

Clara Wang

QSS 30.07

February 21, 2017

Wang 2

I. Synopsis

On December 17, 2010, Mohamed Bouazizi set himself on fire in front of a

municipal office in Tunisia, sparking a wave of protests that toppled governments

across the Middle East. This movement, known as the “Arab Revolution,” has also

been dubbed the “Twitter Revolution” due to the prominent role that social media

played in the rebellions.2 Following these revolutions, authoritarian regimes began

implementing strict information control measures to prevent this phenomenon from

reoccurring. Countries such as China, Vietnam, and North Korea now limit access

to certain websites, 3and Turkey has recently banned Facebook and Twitter in the

wake of an attempted coup.4 These nations have also started flooding the Internet

with propaganda to mitigate the threat of social media to their power. The Chinese

government employs a “Fifty Cent Party” that overwhelms social media with pro-

China posts, fueling nationalistic sentiments across the country.5 Russia has also

created an army of “Internet trolls” that work to influence public opinion both

domestically and abroad. 6 Thus, rebellions have evolved dramatically in the

Internet Age, as both protestors and governments have learned to leverage social

media to organize action and influence opinion. Since most models of rebellion

predate social media, they fail to account for the influence of online social networks.

My model will help explain the role that social media can play in facilitating or

suppressing rebellions.

II. Case Study: The 2014 Ukrainian Revolution and Conflict with Russia

The 2014 Ukrainian revolution, along with the annexation of Crimea, offer a fascinating

case study of the Internet’s role in rebellion and suppression. The revolution, also known as the

“Euromaidan Revolution” or the “Revolution of Dignity,” was triggered by Ukrainian President

Viktor Yanukovych’s November 2013 decision to suspend the Ukraine-European Union

Association Agreement. The public saw the move as a sign that Yanukovych prioritized relations

with Russia over the European Union. To protest his actions, thousands of Ukrainians flooded

Maidan Nezalezhnosti (Independence Square) in Kiev. Although the movement started to die

down in late November, protestors were stirred back into action on November 30, 2013 when the

Berkut Special Police beat students who remained in the square (i.e. the Maidan), introducing

violence into the peaceful protest. One day later, 10,000 people gathered in the square. By

Wang 3

December 1, 2013, around 800,000 people from all over Ukraine had joined the protesters in the

Maidan, demanding that Yanukovych resign.7

Tensions mounted as the protests grew bloodier. Hired thugs, known as “titushkas,”

attacked protestors and journalists without fear of punishment from corrupt police officers, and

the Berkut began loading their guns with actual bullets rather than rubber ones. The violence

climaxed on February 20, 2013, when government snipers killed 67 protestors.8 The protestors

had only been armed with wooden clubs, and they only had shields made from sheet metal or

wood to protect themselves from the bullets. Videos of the massacre saturated the Internet, and

the resulting public outrage led Yanukovych’s parliamentary allies to withdraw their support. On

February 22, 2014, the Verkhovna Rada (Ukraine’s parliamentary body) unanimously voted to

impeach Yanukovych. Shortly afterwards, he packed up his wealth and flew to Russia.9

Unfortunately, celebrations over Yanukovych’s ousting were short lived. Just a week

after Yanukovych fled the country, men in unknown uniforms removed the Crimean prime

minister from power, raised a Russian flag above Crimea’s parliament building, and installed

Sergey Aksyonov as the new prime minister. Aksyonov opposed the new government in Kiev

and called for a referendum vote, which took place on March 16, 2014. Voters were given one of

two options: (1) join Russia or (2) give Crimea sovereign status by returning to the Crimean

Constitution of 1992. Allegedly, 97 percent of voters chose to join Russia.10 Many countries find

these statistics highly suspect, and they view Crimea’s absorption by Russia as an illegal

annexation of Ukrainian territory. Nevertheless, Russia has continued its aggressions towards

Ukraine, as its forces now occupy the Donetsk region of Ukraine. Russia’s tactics for gaining

power and territory in Ukraine have been described as “hybrid warfare,” as their strategy

involves a combination of “soft power” propaganda and “hard power” military force.11

Wang 4

III. Notable Characteristics of Euromaidan and the Ukraine-Russia Conflict

A closer look at the events of Euromaidan and Russian aggression in Ukraine reveals

important features of rebellion and suppression. Below, I identify two, broad characteristics of

these real-world events that will be incorporated into my model of rebellion in the Internet Age.

a) Social Influence Effects

Social influences, particularly from prominent members of society or individuals with

large networks, helped grow the Euromaidan protests. When they first began, the crowds in the

square mainly consisted of students, but then grew to include members from all levels of society

as people reached out to their networks and urged others to join them in the Maidan. Based on

surveys conducted in the midst of the Euromaidan Revolution, 47 percent of people gathered in

the square learned about the protests from their friends, 18 percent from work colleagues, and 15

percent from family members. These social networks were critically important for bringing first-

time protestors to the Maidan. 42 percent of protestors stated that they were prompted to action

from texts sent by a family member or friend.12 Notably, individuals with greater social weight

or status had a greater likelihood of encouraging others to join the protests. For example, the

well-known Ukrainian journalist, Mustafa Nayyem, is popularly credited for bringing the first

crop of protestors to the square by posting the following message on Facebook: “We are meeting

at 22:30 under the Monument of Independence. Dress warm, bring umbrellas, tea, coffee, good

mood and friends. Reposts are highly encouraged!” (translated into English from Ukrainian).13

Another facet of social influence highlighted by the 2014 Ukrainian Revolution is the

relationship between social influence and government legitimacy. Prior to the Euromaidan

Revolution, the Yanukovych regime was broadly considered to be a corrupt, greedy

administration that was embezzling money from the Ukrainian people.14 Thus, it had minimal

Wang 5

legitimacy as a government. Nevertheless, citizens only began actively protesting once social

influences came into play, such as the Facebook post from Mustafa Nayyem. These social

influences then helped grow the protest as people called for others to join them. Hence, social

influence may have helped stabilize the Yanukovych regime’s rule before the protests, even

when government legitimacy was low, as citizens may have observed that no one else spoke out

against the regime so they chose to remain silent as well. However, once a social movement

started to develop, it quickly escalated into a full-blown revolution as social influences created a

“domino effect." Hence, the Euromaidan Revolution reveals that social influence can help

stabilize the status quo, but as government legitimacy drops, this stabilizing effect deteriorates.

The Euromaidan Revolution also revealed that social influence may offset the deterrent

effects of punishment. Throughout the protests, the Yanukovych regime attempted to curtail the

movement by threatening individuals, such as reporters and NGO leaders, as well as increasing

punishments. For example, in mid-January 2014 the government enacted a series of anti-protest

laws with harsh punishments, such as a six-year jail sentence for blocking access to someone’s

residence and a 10 to 15-year sentence for mass disruption. Under the new laws, protestors could

also be arrested for participating in a peaceful gathering while wearing a helmet.15 The

traditional understanding of rebellion suggests that

such punishments may deter others from joining the

movement. However, many Ukrainians chose

instead to join the movement, and protestors began

sporting kitchen colanders and other “helmets” in

defiance of the new anti-protest laws.16 Thus in the

case of Euromaidan, social influence counteracted

the intended pacifying effect of harsh punishments. Figure 2: Maidan protestors wearing "helmets.”

Wang 6

b) Effect of Social Media and the Internet

Similar to the Arab Revolution that shook the world in the early 2000s, the Euromaidan

Revolution has also been described as a revolution driven by social media. Protestors used

platforms such as Facebook, Twitter, and VKontakte (a Russian social network) to amplify the

protests, unite individuals and messages under common themes of the revolution, and coordinate

action. For instance, one witness in the Euromaidan protests, Yevgeny Volokin, stated that

“social media played a part in bringing the events in Odessa to light. At least two web videos live

streamed the initial clashes between pro-Russian and pro-Ukrainian activists and then showed

fighting at the trade union building. Twitter provided photos, updates, and commentary.

Facebook was inundated with postings.”17 Additionally, new social media pages were created to

serve specific needs of the protest, such as coordinating legal support, medical services, and

transportation. A Facebook page, “helpgettomaidan,” was created in early December 2013 to

organize carpools from other citizens and across Kiev to the Maidan.18 Hence, social media

helped catalyze the speed at which protests spread among the public, and it also allowed

protestors to organize and sustain their movement for a long period of time.

Governments have also learned how to leverage social media and the Internet to advance

their positions. As part of their hybrid warfare tactics in the conflict with Ukraine, Russia has

started a pro-Kremlin campaign on VKontakte and Odnoklassniki – the two popular, Russian

social media sites used in Ukraine. Russia pays bloggers and the administrators of popular

VKontakte groups to spread fake news about problems in Ukraine.19 These internet “trolls” post

100 internet comments per day and maintain multiple pro-Kremlin Facebook and Twitter

accounts.20 A well-known example of a fraudulent story propagated by these paid “trolls” is a

report about Galina Pyshnyak, a woman who allegedly witnessed a 3-year old boy being tortured

and crucified by the Ukrainian military in 2014. A video of an interview with Pyshnyak was

Wang 7

widely shared on social media, but her story was later proven false.21 While the effects of

Russian propaganda have had little effect on public support for Russia among Western

Ukrainians, Eastern Ukrainians and residents of Crimea have developed a favorable opinion of

the Kremlin.22 Thus, while the Internet has allowed governments to spread their propaganda to a

broader swathe of people, the effectiveness of this propaganda varies with the recipient’s opinion

of the regime – more legitimate governments have more believable, influential propaganda.

IV. Existing Models of Rebellion

To explain the onset of rebellions such as the 2014 Ukrainian Revolution, scholars have

developed simple models to reflect real-world behavior.

a) Unanticipated Political Revolution (Kuran 1989)

In 1989, Timur Kuran proposed a model to explain the occurrence of unexpected

revolutions.23 His model only has one type of agent, citizens, and they rely on social influences

to determine whether to rebel. Individuals have two types of preferences: public and private.

Public preferences are determined by two (sometimes competing) factors: (A) reputational

utility,i and (B) utility of integrity.ii When the utility of (A) is greater than that of (B), individuals

in Kuran’s model may choose to falsify their public preference. This “preference falsification”

allows for unanticipated revolutions to occur, as individuals hide their personal preference of

rebelling until the value of (B) is greater than that of (A), as described below:

i The utility one gains for having a certain public preference (i.e. social pressure to conform to what everyone else

believes). This is determined by calculating the “collective sentiment” of the public, which is the weighted average

of everyone’s public preferences. Weighting is determined by a person’s degree of social influence. ii This is the utility one gains for expressing his/her preference as the public preference, given his/her private one.

Essentially, it’s the degree of guilt you experience if you act out of accordance with your own beliefs (e.g. if you eat

meat but think it’s unethical to kill animals for consumption).

A = reputational utility (integer from 0-1)

B = utility of integrity (integer from 0-1)

Value of rebelling = A + B

Value of remaining quiet = (1 – A) + (1 – B)

If Value of rebelling > Value of remaining quiet, then individuals rebel.

b) Civil Violence (Epstein 2002)

Joshua Epstein (2002) proposed another model for rebellion where personal grievances

and risk of punishment drive whether individuals choose to protest.24 Epstein’s model contains

two types of actors: “cops” and “agents.” He posited that agents have a certain level of hardship

and a perceived legitimacy of the regime, and these two variables determine that individual’s

level of grievance towards the regime through the following equation:

𝐇 = hardship (integer 0 − 1) 𝐋 = legitimacy of regime (integer 0 − 1) 𝐆 = grievance

𝐆 = 𝐇(1 − 𝐋)

Once an agent’s grievance is high enough, they will consider rebelling depending on their level

of risk aversion. If an agent is risk-averse but observes that many others within their range of

“vision” (i.e. social network) are likely to rebel, which reduces the probability for arrest or

punishment, then risk-averse agents are more likely to join in the rebellion.

𝑹 = 𝑙𝑒𝑣𝑒𝑙 𝑜𝑓 𝑟𝑖𝑠𝑘 𝑎𝑣𝑒𝑟𝑠𝑖𝑜𝑛 (𝑖𝑛𝑡𝑒𝑔𝑒𝑟 0 − 1) 𝑱 = 𝑙𝑒𝑛𝑔𝑡ℎ 𝑜𝑓 𝑗𝑎𝑖𝑙 𝑡𝑒𝑟𝑚

𝑷 = 𝑒𝑠𝑡𝑖𝑚𝑎𝑡𝑒𝑑 𝑎𝑟𝑟𝑒𝑠𝑡 𝑝𝑟𝑜𝑏𝑎𝑏𝑖𝑙𝑖𝑡𝑦iii 𝑸 = 𝑹𝑷J

If 𝑮 − 𝑸 > a certain threshold, 𝑻, an individual will rebel. Otherwise, they will remain quiet.

The “cops” in Epstein’s model arrest active agents within their range of vision, and they never

defect to revolution. Thus, this model presents a situation where decentralized dissidents

(“agents”) come together to start a rebellion in the face of a central authority (“cops”).

iii Calculated from number of other active individuals in that person’s range of “vision” (i.e. in their social network).

Wang 9

Although these two models conceive of rebellion in different ways, they are similar in

that group behavior plays an instrumental role in the onset of a rebellion. In Kuran’s model,

individuals must observe that a certain number of others feel similarly unhappy with the ruling

regime before they are willing to act. In Epstein’s model, individuals are only willing to rebel

when they observe that others are active so that their own risk of arrest is sufficiently low.

V. Critiques of Existing Models

While Kuran’s model offers a simple yet representative model of rebellion, he assumes

that individuals in a society know everyone else’s public preferences – something that is rarely

true. Kuran includes all individual’s public preferences when calculating collective sentiment,

which he uses to determine the reputational utility an individual considers when deciding to

rebel. Although in the real world public opinion polling offers some measure of collective

sentiment for all individuals in society, most people are only familiar with the sentiments of their

personal social network. For example, Girvan and Newman (2002) analyze community

structures and find that most people cluster together in tightly knit groups, and these groups are

only loosely connected to other social networks.25 Since Kuran’s assumption fails to reflect

reality, his model offers a less legitimate explanation for real world behavior.

Epstein’s model offers a logical explanation for rebellion, but it fails to take social

influences into account. The literature on social influence suggests that it can play a significant

role in shaping collective behavior such as protests and revolutions. Frith and Frith (2008)

provided substantial evidence that humans have innate reactions towards others’ behavior. For

example, humans tend to follow one another’s gaze, and the mere present of an ignorant person

in a room can inhibit individual’s ability to complete simple tasks.26 Furthermore, Bikhchandani

et al. (1998) noted that humans exhibit “observational learning” behavior, where they follow the

Wang 10

actions of others around them. The authors specifically cited rebellions as an example of such

behavior, as people were more likely to go out to the square and protest once they observed

others were there.27 Lorenz et al.’s (2011) notion of “information cascades” echo these

phenomena, as they found that if people were provided information about how others had

answered a challenging question, they were more likely to respond in the same way, even if the

answer was wrong.28 Macy (1991) offers further support for the role of social influence in

guiding group behavior, noting that social influences can facilitate coordination and shift the

behavior of an entire group.29 Hence, since social influences have notable effects on human

behavior, Epstein’s model can be improved by incorporating these influences.

Another flaw with Epstein’s model lies in its foundations in the rational-choice model,

which often fails to fully explain human behavior. Kahneman (1988) found multiple violations of

the rational-choice model of behavior in his experiments,30 as he found that the “rationality” of

an individual varied with the context of the situation (2003).31 Goerree and Holt’s (2001)

experiments also suggest that humans often violate the behavioral norms predicted by the

rational-choice model.32 They asked participants to play a number of classic games modeled by

game theory. The rational-choice model predicts that the outcome of these games would fall at

the Nash equilibria, but not all of them did. Hence, their findings suggest that the rational-choice

model may not adequately capture human thinking, meaning that Epstein’s reliance on this

decision-making model fails to accurately reflect human behavior.

Finally, while Kuran (1989) and Epstein’s (2002) models capture important features of

rebellions, they fail to capture some of the complexity introduced in modern-day protests such as

the 2014 Ukrainian Revolution. In today’s age, the Internet and social media have made it easier

to observe others’ behavior, improving information flows and reducing barriers to collective

action. For example, in the 2014 Ukrainian Revolution, protestors used social media to recruit

Wang 11

more protestors and organize themselves. However, governments have responded by banning

access to websites, censoring information, and flooding the Internet with propaganda. Notably,

Russia relies on such efforts to sway Ukrainian public opinion in its favor. Hence, the Internet

adds an interesting new dynamic to the onset of rebellions, as it can be utilized to both facilitate

and suppress protest.

VI. A Model for Rebellion in the Internet Age

To address the flaws with Kuran (1989) and Epstein’s (2002) models, as well as simulate

the role of social media and the Internet in the onset of rebellions, I adapted the “Rebellion”

model from the NetLogo library.33 This model is based off Epstein’s (2002) model of civil

violence. I merged his model with Kuran’s (1989) model of revolution, so that my new model

captured both the rational-choice behavior of Epstein’s agents along with the effects of social

influence described by Kuran’s model. I also added new features to the model to reflect

characteristics of modern-day rebellions that I gleaned from the case study of Ukraine.iv

a) Merging the Two Models

To merge the two models, I meshed together Kuran (1989) and Epstein’s (2002)

mathematical equations for determining when individuals choose to rebel. I took Kuran’s

threshold equation for rebellion:

Individuals rebel if: 𝑉𝑎𝑙𝑢𝑒 𝑜𝑓 𝑟𝑒𝑏𝑒𝑙𝑙𝑖𝑛𝑔 > 𝑉𝑎𝑙𝑢𝑒 𝑜𝑓 𝑟𝑒𝑚𝑎𝑖𝑛𝑖𝑛𝑔 𝑞𝑢𝑖𝑒𝑡

Where the two sides of the equation are defined as:

𝑉𝑎𝑙𝑢𝑒 𝑜𝑓 𝑟𝑒𝑏𝑒𝑙𝑙𝑖𝑛𝑔 = 𝑨 + 𝑩 𝑉𝑎𝑙𝑢𝑒 𝑜𝑓 𝑟𝑒𝑚𝑎𝑖𝑛𝑖𝑛𝑔 𝑞𝑢𝑖𝑒𝑡 = (1 − 𝑨) + (1 − 𝑩)

iv My model can be found in the following Dropbox folder:

https://www.dropbox.com/sh/npz04o2r5ji61ts/AAALsE4tc6hfGyli-40Rln4Ma?dl=0

Wang 12

Where A = reputation utility, and B = utility of integrity. Since B is based off an individual’s

private preference, I replaced B with G – Q from Epstein’s equation, where G = an individual’s

grievance level, and Q = hesitation to rebel due to fear of punishment. This resulted in:

𝑉𝑎𝑙𝑢𝑒 𝑜𝑓 𝑟𝑒𝑏𝑒𝑙𝑙𝑖𝑛𝑔 = 𝑨 + 𝑮 − 𝑸 𝑉𝑎𝑙𝑢𝑒 𝑜𝑓 𝑟𝑒𝑚𝑎𝑖𝑛𝑖𝑛𝑔 𝑞𝑢𝑖𝑒𝑡 = (1 − 𝑨) + (1 − 𝑮 − 𝑸)

To reflect the relationship between social influence and government legitimacy, I changed the

Value of remaining quiet to the following:

𝑉𝑎𝑙𝑢𝑒 𝑜𝑓 𝑟𝑒𝑚𝑎𝑖𝑛𝑖𝑛𝑔 𝑞𝑢𝑖𝑒𝑡 = (0.5𝑳 − 𝑨) + (1 − 𝑮 − 𝑸)

Thus, as legitimacy decreases, the effect of social influence on an individual’s decision to rebel

increases. As legitimacy increases, social influence plays a less substantial role in whether

individuals rebel.

b) Additional Features

In order to reflect propaganda efforts conducted by governments, I added a propaganda

component to the model by including a “propaganda?” switch. When the switch is on, citizens’

grievance levels slightly decrease, but the amount that it drops varies with the legitimacy of the

government. As government legitimacy increases so does the effect of propaganda, as

propaganda campaigns become more believable when the government has some legitimacy.v

However, when government legitimacy is low, citizens are less convinced by propaganda efforts.

To model the influence of the Internet and social media on rebellions, I added two

switches titled “citizen-internet?” and “govt-internet?” If the first switch is on, then citizens can

“see” a greater number of individuals who are active or in jail. Essentially, their social networks

are expanded, and they are susceptible to social influences from a greater number of individuals.

v The relationship between the influence of propaganda and government legitimacy follows a 2x curve.

Wang 13

Turning the switch off can reflect situations where governments block access to the Internet or

certain social media sites, which limits individual’s social networks. If the “govt-internet?”

switch is on, then the government can “see” a greater number of citizens who are active,

allowing them to find more targets to arrest. This feature captures the government’s ability to

monitor citizens over the Internet and punish online activists. Providing the government with

Internet access also increases the effects of the government’s propaganda efforts by 25 percent in

the model, which reflects the influence of pro-government Internet “trolls” on social media.

As demonstrated by the case study of the 2014 Ukrainian Revolution, social influence

can counteract an increased threat of punishment. To capture this behavior in my model, when

citizens assess how many other individuals in their network are active, they count both the active

citizens and 25 percent of previously active citizens who have been jailed. Thus, even when

many activists have been jailed, the deterrent effects of this punishment are weakened, as these

jailed individuals also contribute to social influence effects that push people towards rebellion.

Finally, I took the social weight feature of Kuran’s model and included it in my new

model. I gave each citizen in my model a “social weight,”vi which is used to calculate the

collective sentiment of an individual’s social network. Thus, social weight determines how much

influence individuals in a society exercise on other’s decisions to rebel or remain quiet.

c) Behavior of the Model

If the “social-effects?” toggle is turned off, the model behaves as Epstein conceived of

rebellion. Citizens in the model only consider their level of grievance and the risk of punishment

when deciding whether to rebel. If the “govt-legitimacy” slider is set to about 0.65, the citizens

in the model fluctuate between rebelling and remaining quiet. When citizens rebel, once a certain

vi The social weight is a turtle attribute and is a random integer between 0 and 1.

Wang 14

number of individuals have been jailed, citizens tend to quiet down again because the threat of

punishment has increased.

However, if the “social-effects?” toggle is turned on, allowing for the new model

incorporating social influences to operate, social effects help stabilize the behavior of citizens. If

“govt-legitimacy” is set to 0.65, rather than fluctuating between periods of quiet and rebellion,

citizens remain quiet overall with very few individuals rebelling. But, if government legitimacy

decreases and a certain threshold is reached, society rapidly explodes into sudden, widespread

protest – exactly what Kuran attempted to reflect in his model based on social influences. As

government legitimacy increases, the shift back to being quiet is quick and sudden as well. Thus,

the model demonstrates the catalyzing effect that social influence can have on shifts to rebellion

or quiet acquiescence to government control, as well as the stabilizing effects of social influence.

Toggling “propaganda?” on or off changes the threshold at which society shifts from

quiet to rebellious. Since propaganda decreases citizens’ level of grievance, the threshold at

which society erupts into protest falls at a lower level of government legitimacy. The toggles for

“govt-internet” and “citizen-internet” also affect this threshold. Turning “govt-internet” on

increases the risk of arrest and the number of individuals in jail, meaning that the threshold for

protest drops to a lower government legitimacy. Turning “citizen-internet” on increases social

influence effects, so sudden shifts to protest or peace occur even faster. Allowing citizens to

access the internet also leads to more stable states of society; peaceful or rebellious states last

much longer when citizens are susceptible to the effects of larger, online social networks. Hence,

the Internet can be used as both a rebellion-inducing or suppressive force.

Wang 15

d) Important Considerations

To simplify my model so that it was feasible to create in NetLogo, I made a few key

assumptions:

1. I assumed that the social weight of each citizen varied randomly, and that they

exerted the same degree of influence on every other citizen in the model. Hence, I did

not consider that one person may have substantial influence on one citizen (e.g. a

mother on her child), but less of an influence on another citizen (e.g. the same mother

and her gym instructor).

2. I assumed that the Internet caused the same degree of increase for the government and

citizens’ networks (an increase in radius of five patches). I did not vary the degree of

increase across individuals, even though in reality some individuals may have

substantially larger networks on the Internet (e.g. celebrities have many social media

followers, while average citizens generally have less).

VII. Conclusion

In essence, this model considers two schools of thought about how humans behave:

rational-choice and social influence. The rational-choice model of decision-making is captured

by the cost-benefit analysis individuals conduct when they are assessing the risks of joining a

rebellion, or the social costs they incur for failing to align their preferences with public opinion.

The social influence model can be seen in the way individuals follow one another’s behavior,

leading to sudden, widespread shifts to protest or quiescence, as well as sustained and stable

states of society. As the rational-choice model often fails to fully explain human behavior,

adding this nuance of social influence may capture how humans respond to propaganda, and how

the Internet has affected the onset of rebellions.

Wang 16

As previously referenced, the 2014 Ukrainian Revolution stands as a fitting example for

this model. Although government legitimacy was low under the Yanukovych regime, society

remained in a stable state of quiescence, possibly due to both rational assessment of risks as well

as social influence factors. However, once influential members of society encouraged protestors

to gather in Independence Square, the country erupted in revolution. Social media helped

facilitate and sustain the revolution, and these catalyzing effects can be seen in the model.

However, in the case of Ukraine the Internet has helped empower regimes as well – namely the

Kremlin. Utilizing social media networks, Russians have leveraged social influence effects to

sway Ukrainians against their own government. These efforts have proven successful, as

demonstrated by Russia’s annexation of Crimea and the popular support that Russia experiences

in Eastern Ukraine.

Thus, the Internet has become a powerful tool for both protest and suppression alike. By

taking advantage of the natural effects of social influence, citizens and governments can work to

facilitate rebellion or quell the masses. This past year, the New York Times published an article

proclaiming the “globe-shaking” power of social media.34 Perhaps a more precise description

might be the “globe-shaking” power of magnified social influence. As suggested by the situation

in Ukraine, as well as the behavior of my proposed model, the Internet has enhanced the effects

of social influence to the point where it may play a far more significant role than rational-choice

decision-making when it comes to collective behavior.

Wang 17

1 Usree Bhattacharya, “Revolutionary Twitter,” Found in Translation, 11 Aug. 2009,

http://foundintranslation.berkeley.edu/?p=4638. 2 Mariam Esseghaier, “Tweeting Out a Tyrant: Social media and the Tunisian Revolution,” Journal of Mobile Media

6, no. 3 (2012), http://wi.mobilities.ca/tweeting-out-a-tyrant-social-media-and-the-tunisian-revolution/. 3 The Committee to Protect Journalists, “10 Most Censored Countries,” CPJ, last updated 2015, accessed on 1 Feb.

2017, https://www.cpj.org/2015/04/10-most-censored-countries.php. 4 May Bulman, “Facebook, Twitter and Whatsapp blocked in Turkey after arrest of opposition leaders,”

Independent, 4 Nov. 2016, http://www.independent.co.uk/news/world/asia/facebook-twitter-whatsapp-turkey-

erdogan-blocked-opposition-leaders-arrested-a7396831.html. 5 Henry Farrell, “The Chinese government fakes nearly 450 million social media comments a year. This is why.”

The Washington Post, 19 May 2016, https://www.washingtonpost.com/news/monkey-cage/wp/2016/05/19/the-

chinese-government-fakes-nearly-450-million-social-media-comments-a-year-this-is-

why/?utm_term=.efb52847c168. 6 Leo Benedictus, “Invasion of the troll armies: from Russian Trump supporters to Turkish state stooges,” The

Guardian, 6 Nov. 2016, https://www.theguardian.com/media/2016/nov/06/troll-armies-social-media-trump-russian. 7 Borislaw Bilash II, “How it all happened,” Euromaidan Press, 20 Feb. 2016,

http://euromaidanpress.com/2016/02/20/the-story-of-ukraine-starting-from-euromaidan/2/. 8 Bilash II. 9 “Ukrainian MPs vote to oust President Yanukovych,” BBC News, 22 Feb. 2014, http://www.bbc.com/news/world-

europe-26304842. 10 “Ukraine crisis: Timeline,” BBC News, 13 Nov. 2014, http://www.bbc.com/news/world-middle-east-26248275. 11 Bilash II. 12 Olga Onuch, “Social networks and social media in Ukrainian “Euromaidan” protests,” The Washington Post, 2

Jan. 2014, https://www.washingtonpost.com/news/monkey-cage/wp/2014/01/02/social-networks-and-social-media-

in-ukrainian-euromaidan-protests-2/. 13 Tetyana Bohdanova, “Unexpected revolution: the role of social media in Ukraine’s Euromaidan uprising,”

European View 13, no. 1 (June 2014): 133, doi: 10.1007/s12290-014-0296-4. 14 “Ukraine crisis: Yanukovych and the tycoons.” BBC News, 11 Dec. 2013, http://www.bbc.com/news/world-

europe-25323964. 15 “Kolesnychenko-Oliynyk laws: infographics,” Euromaidan Press, 12 Jan. 2014,

http://euromaidanpress.com/2014/01/12/kolesnychenko-oliynyk-laws-infographics/. 16 Caitlin Dewey, “Why some Ukrainian protestors are wearing kitchen colanders,” The Washington Post, 22 Jan.

2014, https://www.washingtonpost.com/news/worldviews/wp/2014/01/22/why-some-ukrainian-protesters-are-

wearing-kitchen-colanders/?utm_term=.ba0da3b33323. 17 David Stern, “The Twitter War: Social Media’s Role in Ukraine Unrest,” National Geographic, 11 May, 2014,

http://news.nationalgeographic.com/news/2014/05/140510-ukraine-odessa-russia-kiev-twitter-world/. 18 Joshua A. Tucker, Megan Metzger, Duncan Penfold-Brown, Richard Bonneau, John Jost, and Jonathan Nagler,

“Protest and Social Media: Technology and Ukraine’s #Euromaidan,” Carnegie Reporter 7, no. 4 (2014): 8,

https://www.carnegie.org/media/filer_public/83/c2/83c2aac2-0089-4426-b6c3-

7f4e71c3f13a/ccny_creporter_2014_vol7no4.pdf. 19 Daniella Peled, “Ukraine’s Social Media Revolution,” Institute for War & Peace Reporting, last updated 28 Mar.

2014, accessed on 8 Oct. 2016, https://iwpr.net/global-voices/ukraines-social-media-revolution. 20 “Analysis of Russia’s Information Campaign Against Ukraine,” NATO StratCom Centre of Excellence, 2014,

accessed on 6, Oct. 2016, http://www.stratcomcoe.org/analysis-russias-information-campaign-against-ukraine. 21 “Analysis of Russia’s Information Campaign Against Ukraine,” NATO StratCom Centre of Excellence, 2014,

accessed on 6, Oct. 2016, http://www.stratcomcoe.org/analysis-russias-information-campaign-against-ukraine. 22 Kenneth Rapoza, “One Year After Russia Annexed Crimea, Locals Prefer Moscow to Kiev,” Forbes, 20 Mar.

2015, http://www.forbes.com/sites/kenrapoza/2015/03/20/one-year-after-russia-annexed-crimea-locals-prefer-

moscow-to-kiev/2/#35681f652a9d. 23 Timur Kuran, “Sparks and prairie fires: A theory of unanticipated political revolution,” Public Choice 61, no. 1

(1989) 61: 41, doi:10.1007/BF00116762. 24 Joshua M. Epstein, “Modeling Civil Violence: An Agent-Based Computational Approach,” Proceedings of the

National Academy of Sciences 99, no. Supplement 3 (May 14, 2002): 7243–50, doi:10.1073/pnas.092080199. 25 M. Girvan and M.E.J. Newman, “Community structure in social and biological networks,” PNAS 99, no. 12

(2002): 7821-26, doi: 10.1073/pnas.122653799.

Wang 18

26 Chris D. Frith and Uta Frith, “Implicit and Explicit Processes in Social Cognition,” Neuron 60 (2008): 503-510,

doi: 10.1016/j.neuron.2008.10.032. 27 Sushil Bikhchandani, David Hirshleifer, and Ivo Welch, “Learning from the Behavior of Others: Conformity,

Fads, and Informational Cascades,” The Journal of Economic Perspectives 12, no. 3 (1998): 151-170,

http://www.jstor.org/stable/2647037. 28 Jan Lorenz, Heiko Rauhut, Frank Schweitzer, and Dirk Helbing, “How social influence can undermine the

wisdom of crowd effect,” PNAS (2011): 1-6, doi:10.1073/pnas.1008633108. 29 Michael W. Macy, “Chains of Cooperation: Threshold Effects in Collective Action,” American Sociological

Review 56, no. 6 (1991): 730-747, http://links.jstor.org/sici?sici=0003-

1224%28199112%2956%3A6%3C730%3ACOCTEI%3E2.0.CO%3B2-Y. 30 Daniel Kahneman, “Experimental Economics: A Psychological Perspective,” in Lecture Notes in Economics and

Mathematical Systems, eds. M. Beckmann and W. Krelle, Springer-Verlag (Berlin): 1989. 31 Daniel Kahneman, “A Psychological Perspective on Economics,” The American Economic Review 93, no. 2

(2003): 162-168, http://www.jstor.org/stable/3132218. 32 Jacob K. Goeree and Charles A. Holt, “Ten Little Treasures of Game Theory and Ten Intuitive Contradictions,”

The American Economic Review 91, no. 5 (2001): 1402-1422, http://www.jstor.org/stable2677931. 33 Uri Wilensky, (2004), “NetLogo Rebellion model,” http://ccl.northwestern.edu/netlogo/models/Rebellion, Center

for Connected Learning and Computer-Based Modeling, Northwestern University, Evanston, IL. 34 Farhad Manjoo, “Social Media’s Globe-Shaking Power,” the New York Times, 16 Nov. 2016,

https://www.nytimes.com/2016/11/17/technology/social-medias-globe-shaking-power.html?smid=tw-

nytimes&smtyp=cur&_r=0.

Wang 19

Works Cited

“Analysis of Russia’s Information Campaign Against Ukraine.” NATO StratCom Centre of

Excellence. 2014. Accessed on 6, Oct. 2016. http://www.stratcomcoe.org/analysis-

russias-information-campaign-against-ukraine.

Benedictus, Leo. “Invasion of the troll armies: from Russian Trump supporters to Turkish state

stooges.” The Guardian. 6 Nov. 2016.

https://www.theguardian.com/media/2016/nov/06/troll-armies-social-media-trump-

russian.

Bhattacharya, Usree. “Revolutionary Twitter.” Found in Translation. Last updated 11 Aug.

2009. Accessed on 1 Feb. 2017. http://foundintranslation.berkeley.edu/?p=4638.

Tetyana Bohdanova, “Unexpected revolution: the role of social media in Ukraine’s Euromaidan

uprising,” European View 13, no. 1 (June 2014): 133, doi: 10.1007/s12290-014-0296-4.

Bulman, May. “Facebook, Twitter and Whatsapp blocked in Turkey after arrest of opposition

leaders.” The Independent. 4 Nov. 2016.

http://www.independent.co.uk/news/world/asia/facebook-twitter-whatsapp-turkey-

erdogan-blocked-opposition-leaders-arrested-a7396831.html.

Bikhchandani, Sushil, David Hirshleifer, and Ivo Welch. “Learning from the Behavior of Others:

Conformity, Fads, and Informational Cascades.” The Journal of Economic Perspectives

12, no. 3 (1998): 151-170. http://www.jstor.org/stable/2647037.

Bilash II, Borislaw. “How it all happened.” Euromaidan Press. 20 Feb. 2016.

http://euromaidanpress.com/2016/02/20/the-story-of-ukraine-starting-from-

euromaidan/2/.

The Committee to Protect Journalists. “10 Most Censored Countries.” CPJ. Last updated 2015.

Accessed on 1 Feb. 2017. https://www.cpj.org/2015/04/10-most-censored-countries.php.

Wang 20

Dewey, Caitlin. “Why some Ukrainian protestors are wearing kitchen colanders.” The

Washington Post. 22 Jan. 2014.

https://www.washingtonpost.com/news/worldviews/wp/2014/01/22/why-some-ukrainian-

protesters-are-wearing-kitchen-colanders/?utm_term=.ba0da3b33323.

Epstein, Joshua M. “Modeling Civil Violence: An Agent-Based Computational Approach.”

Proceedings of the National Academy of Sciences 99, no. Supplement 3 (May 14, 2002):

7243–50. doi:10.1073/pnas.092080199.

Esseghaier, Mariam. “Tweeting Out a Tyrant: Social media and the Tunisian Revolution.”

Journal of Mobile Media 6, no. 3 (2012). http://wi.mobilities.ca/tweeting-out-a-tyrant-

social-media-and-the-tunisian-revolution/.

Farrell, Henry. “The Chinese government fakes nearly 450 million social media comments a

year. This is why.” The Washington Post. 19 May 2016.

https://www.washingtonpost.com/news/monkey-cage/wp/2016/05/19/the-chinese-

government-fakes-nearly-450-million-social-media-comments-a-year-this-is-

why/?utm_term=.efb52847c168.

Frith, Chris D. and Uta Frith. “Implicit and Explicit Processes in Social Cognition.” Neuron 60

(2008): 503-510. doi: 10.1016/j.neuron.2008.10.032.

Girvan, M. and M.E.J. Newman. “Community structure in social and biological networks.”

PNAS 99, no. 12 (2002): 7821-26. doi: 10.1073/pnas.122653799.

Goeree, Jacob K., and Charles A. Holt. “Ten Little Treasures of Game Theory and Ten Intuitive

Contradictions.” The American Economic Review 91, no. 5 (2001): 1402-1422.

http://www.jstor.org/stable2677931.

Wang 21

Kahneman, Daniel. “Experimental Economics: A Psychological Perspective.” in Lecture Notes

in Economics and Mathematical Systems. Eds. M. Beckmann and W. Krelle. Springer-Verlag

(1989): Berlin.

Kahneman, Daniel. “A Psychological Perspective on Economics.” The American Economic

Review 93, no. 2 (2003): 162-168. http://www.jstor.org/stable/3132218.

“Kolesnychenko-Oliynyk laws: infographics,” Euromaidan Press, 12 Jan. 2014,

http://euromaidanpress.com/2014/01/12/kolesnychenko-oliynyk-laws-infographics/.

Kuran, Timur. “Sparks and prairie fires: A theory of unanticipated political revolution.” Public

Choice 61, no. 1 (1989): 41. doi:10.1007/BF00116762.

Lorenz, Jan, Heiko Rauhut, Frank Schweitzer, and Dirk Helbing. “How social influence can

undermine the wisdom of crowd effect.” PNAS (2011): 1-6.

doi:10.1073/pnas.1008633108.

Macy, Michael W. "Chains of cooperation: Threshold effects in collective action." American

Sociological Review (1991): 730-747.

Manjoo, Farhad. “Social Media’s Globe-Shaking Power.” the New York Times. 16 Nov. 2016.

https://www.nytimes.com/2016/11/17/technology/social-medias-globe-shaking-

power.html?smid=tw-nytimes&smtyp=cur&_r=0.

Onuch, Olga. “Social networks and social media in Ukrainian “Euromaidan” protests.” The

Washington Post. 2 Jan. 2014, https://www.washingtonpost.com/news/monkey-

cage/wp/2014/01/02/social-networks-and-social-media-in-ukrainian-euromaidan-

protests-2/.

Peled, Daniella. “Ukraine’s Social Media Revolution.” Institute for War & Peace Reporting.

Last updated 28 Mar. 2014. Accessed on 8 Oct. 2016. https://iwpr.net/global-

voices/ukraines-social-media-revolution.

Wang 22

Rapoza, Kenneth. “One Year After Russia Annexed Crimea, Locals Prefer Moscow to Kiev.”

Forbes. 20 Mar. 2015. http://www.forbes.com/sites/kenrapoza/2015/03/20/one-year-

after-russia-annexed-crimea-locals-prefer-moscow-to-kiev/2/#35681f652a9d.

Stern, David. “The Twitter War: Social Media’s Role in Ukraine Unrest.” National Geographic.

11 May, 2014. http://news.nationalgeographic.com/news/2014/05/140510-ukraine-

odessa-russia-kiev-twitter-world/.

Tucker, Joshua A., Megan Metzger, Duncan Penfold-Brown, Richard Bonneau, John Jost, and

Jonathan Nagler. “Protest and Social Media: Technology and Ukraine’s #Euromaidan.”

Carnegie Reporter 7, no. 4 (2014): 8.

https://www.carnegie.org/media/filer_public/83/c2/83c2aac2-0089-4426-b6c3-

7f4e71c3f13a/ccny_creporter_2014_vol7no4.pdf.

Wilensky, Uri. (2004). NetLogo Rebellion model.

http://ccl.northwestern.edu/netlogo/models/Rebellion. Center for Connected Learning and

Computer-Based Modeling, Northwestern University, Evanston, IL.

“Ukraine crisis: Timeline.” BBC News. 13 Nov. 2014. http://www.bbc.com/news/world-middle-

east-26248275.

“Ukrainian MPs vote to oust President Yanukovych.” BBC News. 22 Feb. 2014.

http://www.bbc.com/news/world-europe-26304842.